|
|
||||||||
Department of Animal Sciences, University of Illinois, Urbana 61801
| Abstract |
|---|
|
|
|---|
Key Words: Estradiol Involution Lactation Mammary Glands Sows
| Introduction |
|---|
|
|
|---|
In swine, the gland is still actively growing during the lactation period (Kim et al., 1999
) in response to milk removal by the piglets (Hurley, 2001
). Cessation of milk removal typically occurs by abrupt removal of the litter during the third or fourth week of lactation. Little secretion is present by d 8 after weaning (Cross et al., 1958
), and secretory compositional changes are complete within the first week after weaning (Atwood and Hartmann, 1995
). Little further information is available about the process of mammary gland involution in swine. In addition, whereas estrogen has involution-inducing properties in ruminants during late lactation, its effects on the rate of mammary gland involution in swine are unknown. The objectives of this study were to characterize the tissue compositional changes in sow mammary glands after weaning and to determine if estradiol administration alters the profile of tissue changes.
| Materials and Methods |
|---|
|
|
|---|
Sows were randomly assigned to one of two treatment groups after weaning: nonestrogen-treated control and estrogen treated. Estradiol-17ß (Sigma Chemical Co., St. Louis, MO) was administered as twice daily i.m. injections of 0.125 mg of estradiol-17ß kg of BW. Estradiol-17ß was dissolved in ethanol and emulsified in peanut oil as carrier. Control sows received the peanut oil vehicle. Sows were killed either on d 0 (d of weaning; n = 5) or on d 2, 3, 4, 5, or 7 after weaning (n = 4 for each treatment on each day). Sows were slaughtered at the University of Illinois Meat Science Laboratory (Urbana, IL). The protocol for this experiment was approved by the Institutional Animal Care and Use Committee of the University of Illinois at Urbana-Champaign.
Tissue Collection
For each scheduled slaughter, sows were transported to the University of Illinois Meat Science Laboratory at 0600 before the morning feeding. Sows were stunned electrically and killed by exsanguination. All mammary glands were removed from the sows and the skin and extraneous fat pad were dissected from parenchymal tissue. Suckled and nonsuckled glands were identified based on the teat order observation. Individual mammary glands were dissected, weighed, and bisected in an approximate mid-sagittal section to measure cross-sectional area. Cross-sectional area was determined by tracing the outline of the parenchymal tissue on the cut-face of each gland onto a transparency film and the area was measured using a mechanical polar planimeter (LASICO L-10; Los Angeles Scientific Instrument Co., Inc., Los Angeles, CA). Half of each gland was ground in a commercial blender (Waring Products, New Hartford, CT) and stored at -20°C for chemical analysis (Kim et al., 1999
).
Chemical Analyses
Frozen ground tissue was used for measuring tissue DM content, fat content, protein content, ash content, and DNA content. Dry matter content of tissue was measured by desiccation at 105°C for 8 h. Crude fat content was determined by Soxhlet extraction of previously dried tissue using a chloroform:methanol (87:13) binary extracting solution (Novakofski et al., 1989
). Tissue protein content was obtained by measuring nitrogen content with the Kjeldahl method (AOAC, 1995
). Ash content was measured by combustion of dried tissue at 450°C for 8 h. Measurement of DNA content was performed in triplicate according to a modification of the method of Labarca and Paigen (1980)
, as described previously (Kim et al., 1999
). Briefly, approximately 0.05 g of wet mammary tissue was homogenized with 2 mL of homogenization buffer (10 mM Tris base, 5 mM EDTA, and 0.5% cholamidopropyl-dymethylammonio-propane-sulfonate [Sigma Chemical Co.], pH 7.2). Just before homogenization, the protease inhibitors, phenylmethylsulfonyl-fluoride, and
-amino caproic acid (Sigma Chemical Co.) were added at concentrations of 0.1 mM. After homogenization, samples were sonicated for 15 s (Tekmar Sonic Disruptor, Tekmar Co., Cincinnati, OH).
Statistical Analyses
Data from mammary glands that were known to have been suckled throughout the lactation period were analyzed separately from data for glands that were not suckled throughout the lactation period. Data from mammary glands obtained at weaning (d 0) were used to establish a baseline of mammary gland characteristics at the end of lactation and therefore were included in the results. Mean responses were modeled to investigate the effect of day of involution on the mammary gland. Statistical analyses of the data were performed using the MIXED procedure of SAS (SAS Inst., Inc., Cary, NC). Data were analyzed by treatment group (control and estrogen treated), and across treatment. In all analyses, day of involution was included as a factor and as regressor (with linear and quadratic terms) to gain complementary insights on the trends across time. The model is as follows:
![]() |
where yijkl is the response variable (e.g., cross-sectional area), µ is the overall population mean, ai is the effect of the ith estrogen treatment group (fixed effect), bj is the effect of the jth day of involution (fixed effect), cijk is the random effect of sow k nested within treatment level i and day j, and eijkl is the error term. The assumptions are that the sow effects are independent and normally distributed with equal variance and that the residuals are independent and normally distributed with equal variance and both random effects are independent. The differences among factor levels were evaluated based on P-values, least-square means and standard errors.
| Results |
|---|
|
|
|---|
|
|
Changes in mammary parenchymal tissue in untreated sows after weaning (Table 1
) also included a decline in dry tissue weight (linear effect, P < 0.0001), gland protein weight (linear effect, P < 0.0001; quadratic effect, P = 0.015), gland fat weight (linear effect, P < 0.0001), and gland ash weight (linear effect, P < 0.0001; quadratic effect, P = 0.0128).
The proportion of DNA per mass of wet tissue did not change after weaning (Table 2
), suggesting that the number of cells per milligram of wet tissue did not change during the involution process. However, the composition of mammary parenchymal tissue did change after weaning. Dry tissue as a percentage of wet tissue weight increased (linear effect, P = 0.0111; quadratic effect, P = 0.0283) between d 0 and 7, with the major increase occurring between d 0 and d 2 (Table 2
). The protein percentage of dry tissue declined after weaning (linear effect, P = 0.0032; quadratic effect, P = 0.0098), as did the ash percentage of dry tissue (linear effect, P < 0.0001; quadratic effect, P = 0.0225). Conversely, fat percentage of dry tissue increased after weaning (linear effect, P = 0.0003; quadratic effect, P = 0.0022). At least some of the increased proportion of dry tissue during the overall involution of the gland must have been accumulation of lipid in the remaining tissue. Indeed, fat percentage of dry tissue was positively correlated with dry tissue as a percentage of wet tissue weight (r = 0.84; P < 0.0001) but negatively correlated with protein percentage of dry tissue (r = -0.80; P < 0.0001) after weaning.
|
Nonsuckled Glands
No significant effects of day after weaning or day x treatment interaction were found for any variable measured on mammary parenchymal tissue after weaning for the glands that had not been suckled during lactation. Overall means for the 7 d after weaning are provided in Table 3
. This lack of change in tissue morphology also is evident in the gross structure of the excised parenchymal tissue from nonsuckled glands (Figure 1
).
|
| Discussion |
|---|
|
|
|---|
Piglets typically are weaned between 14 and 21 d after farrowing. This abrupt cessation of milk removal causes initiation of involution in the sows mammary glands. Turner (1952)
noted that secretory activity continued for 24 to 36 h after weaning, as indicated by gland size. Histological examinations of weaned mammary glands indicate that the mammary gland initially becomes engorged through the first day after weaning, but it then appears as though milk in the lumen begins to be reabsorbed (Cross et al., 1958
). Furthermore, the greatest magnitude of change in gland mass and DNA occurs between d 0 of involution (day piglets are removed from the sow) and d 2 after weaning. The extensive loss in mass and cells during the initial 7 d after weaning are indicative of the rapidly changing physiology of the gland during this period of regression.
The abrupt weaning of piglets at peak lactation induces the sows mammary glands to begin a period of active involution. The sows mammary glands seem to pass through three phases during the initial week after weaning. These phases approximately coincide with periods between the day of weaning to 2 d after weaning, 2 d after weaning to 4 or 5 d after weaning, and d 4 or 5 after weaning to d 7 after weaning.
The initial phase of mammary gland involution in the sow begins when the piglets are weaned. In the absence of milk removal, further milk secretion is inhibited by accumulation of an autocine factor, the feedback inhibitor of lactation (Wilde et al., 1995
; Knight et al., 1998
). Frequent suckling removes this autocrine factor. In lactating sows, milk refilling time in the glands is estimated at approximately 35 min (Spinka et al. 1997
). When piglets are removed from the sow, the mammary glands quickly refill with milk, the feedback inhibitor of lactation accumulates in the alveolar lumen, and further milk secretion is inhibited. In the case of the sow that has just had a litter weaned, removal of the litter for as little as 2 h results in a significant decrease in prolactin concentration in the sow (Bevers et al., 1978
).
By 6 h after removal of the litter, alveoli are moderately distended and further distention has occurred by 13 h after weaning (Cross et al., 1958
). However, by 24 h after weaning there are few alveoli that remain distended, and by 48 h, little fluid volume remains in the alveoli (Cross et al., 1958
), suggesting that a loss of tissue fluid occurs during this period. Results from the present study also indicate a loss of tissue fluid and of other tissue components. In the first 2 d after weaning, mammary parenchymal cross-sectional area shrinks by over 25%. This increment of decline represents almost half of the reduction in mammary gland parenchymal cross-sectional area observed during all of the 7 d after weaning. There are parallel declines in mammary gland wet weight, DNA, dry weight, protein weight, and ash weight. The loss of water in the mammary parenchymal tissue also is indicated by the nearly 30% increase in gland dry tissue percentage in the initial 2 d after weaning. This increase in gland dry tissue percentage is directly related to the loss of water in the gland and is accompanied by the loss of mammary gland wet weight. The mass of fat present in the parenchymal tissue increases during the initial 2 to 3 d after weaning. In addition to the increase of fat weight, there is an increase in fat percentage in dry tissue and a corresponding decrease in protein percentage. This apparent increase in tissue fat content during early involution may reflect a transitory accumulation of milk lipid in epithelial cells of the gland observed histologically by Cross et al. (1958)
.
The extensive loss of mammary parenchymal DNA during involution indicates a loss of cells in the gland. In the mouse mammary gland, tissue morphology consistent with apoptosis is observed within 2 d of milk stasis (Strange et al., 1992
; Walker et al., 1989
), and apoptosis is responsible for the loss of epithelial cells during involution (Li et al., 1997
). This cell loss is extensive in the mouse, causing disintegration of the alveolar structures during the initial stages of involution. Milk stasis in dairy animals also stimulates apoptosis in mammary tissue (Quarrie et al., 1994
; Wilde et al., 1997
; Wilde et al., 1999
), although the loss of structure is not as severe as in the mouse. The present study demonstrates that the sow mammary gland undergoes considerable cell loss in early involution, presumably occurring through apoptosis.
Following the initial large decrease in mammary gland mass in the first 2 d after weaning, there is a period of more limited additional change in mammary gland component mass from d 2 to 3 or 4. This is consistent with the limited change in milk metabolites observed from the second through the fourth day after weaning (Atwood and Hartmann, 1995
). During this time, it usually is still possible to remove some mammary secretion from the sow with the use of oxytocin (Atwood and Hartmann, 1995
; J. A. Ford and W. L. Hurley, unpublished observation). Although mass of tissue components decline slowly during this period, the composition of milk remaining in the gland changes substantially. Lactose concentration undergoes its greatest decline between d 2 and 4 after weaning, and milk sodium concentrations increase between d 2 and 4 (Atwood and Hartmann, 1995
). A rise in sodium concentration in the mammary secretion indicates that a disruption of tight junctions between epithelial cells is occurring during this time.
The third phase of active involution corresponds to d 5 after weaning through at least d 7 after weaning. The limited mammary secretions that can be collected after d 5 are extremely viscous (Atwood and Hartmann, 1995
). Histological examination of mammary tissue on d 4 after weaning reveals little alveolar structure remaining (Cross et al., 1958
). At 8 d after weaning, there are no signs of alveolar cells in the sow mammary gland (Cross et al., 1958
). Histologically, mammary gland involution appears to be complete by d 8. Results from the present study indicate that during this third phase of mammary gland involution, there is a further decline in mammary gland parenchymal mass, as well as in DNA content and cross-sectional area.
Glands that are not regularly suckled after farrowing undergo regression within the first 7 to 10 d of lactation (Kim et al., 2001
). The pattern of regression of these nonsuckled glands during early lactation, as described by Kim et al. (2001)
, is similar to the pattern observed in the present study for the postweaning involution of the glands that are suckled throughout lactation. In the initial 7 d after weaning, the loss of about two-thirds of the total wet weight mass of glands that are suckled during lactation is comparable with the proportional loss of mass in the nonsuckled glands that occurs in early lactation. Glands that are not suckled during lactation do not seem to undergo further loss of tissue mass during the period after weaning. It is interesting to compare the parenchymal tissue wet weight of glands not suckled during lactation after weaning (averaging approximately 80 g; Table 3
) with the d-7 postweaning wet weight of glands that had been suckled during lactation (averaging approximately 150 g; Table 1
). Glands that are suckled during lactation undergo substantial growth, approximately doubling in size (Kim et al., 1999
), and therefore start postweaning involution at a larger size. Although we cannot rule out further loss of mammary tissue in previously suckled glands after 7 d postweaning, the retention of greater mass of tissue in previously suckled glands may contribute to greater productivity of those glands in the subsequent lactation. In fact, glands that are suckled in a first lactation have enhanced productivity in the next lactation when compared with glands that are not suckled in the first lactation (Fraser et al., 1992
).
In the dairy cow, the rate of mammary gland involution is stimulated by administration of estrogen (Athie et al., 1996
). The elevated blood concentration of estrogen found during late pregnancy is thought to contribute to decreased milk yield and relate to the faster involution observed in pregnant cows during late lactation (Bachman et al., 1988
; Athie et al., 1996
). In the dairy ruminant, mammary gland involution usually occurs during the declining phase of lactation when substantial cell loss is already occurring (Knight and Peaker, 1984
; Wilde and Knight, 1989
). In contrast, the typical 21-d lactation of a sow occurs exclusively during a period of mammary gland growth (Kim et al., 1999
; Hurley, 2001
). Although estrogen is an important mammogenic hormone in swine (Winn et al., 1994
), suckling and milk removal are the major stimulators of mammary growth during lactation in the sow (Hurley, 2001
). Results from the present study indicate that estrogen administration during the postweaning period does not enhance overall rate of mammary involution in sows when litters are weaned at 21 d of lactation.
| Implications |
|---|
|
|
|---|
| Footnotes |
|---|
2 Present address: Dept. of Anim. and Food Sci., Texas Tech University, Lubbock 79409. ![]()
3 Correspondence: 1207 W. Gregory Dr. (fax: 217-333-8804; E-mail: wlhurley{at}uiuc.edu).
Received for publication November 2, 2002. Accepted for publication June 17, 2003.
| Literature Cited |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. A. Ford Jr., S. G. Clark, E. M. Walters, M. B. Wheeler, and W. L. Hurley Estrogenic effects of genistein on reproductive tissues of ovariectomized gilts J Anim Sci, April 1, 2006; 84(4): 834 - 842. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |